An Introduction to Karl Marx

i

An Introduction to Karl Marx

iiiii

An Introduction to Karl Marx

JON ELSTER

The right of the University of Cambridge to print and sell all manner of books was granted by Henry VIII in 1534. The University has printed and published

continuously since 15874.

CAMBRIDGE UNIVERSITY PRESS

Cambridge

London New York New Rochelle

Melbourne Sydney

iv

Published by the Press Syndicate of the University of Cambridge

The Pitt Building, Trumpington Street, Cambridge CB2 1RP 32 East 57th Street, New York, NY 10022, USA

10 Stamford Road, Oakleigh, Melbourne 3166, Australia

Jon Elster First published 1986

Printed in the United States of America

Library of congress cataloging to karl marx.

Elster, Jon, 1940-

An introduction to Karl Marx.

Includes bibliographies.

1. Marx, Karl, 1818-1883. I. Title.

HX39.5.E43 1986 335.4 86-4238

ISBN 0 521 32922 1 hard covers

ISBN 0 521 33831 X paperback

v

Contents

Preface VII

1. Overview 1

Introduction 1

Marx: Life and Writings 5

Marx and Engels , 11

Marxism after Marx 12

Editions of Marx's Writing 17

Bibliography 19

2. Marxist Methodology 21

Introduction 21 Methodological Individualism 22

Marxism and Rational Choice 25

Functional Explanation in Marxism 3 1

Dialectics 34

Bibliography 39

3. Alienation 41

Introduction

Alienation: Lack of Self-realization 43

Alienation: Lack of Autonomy 49

Alienation: The Rule of Capital over Labor 54

Fetishism 56

Bibliography 58

4. Marxian Economics 60

Introduction 60

The Labor Theory of Value 63

Reproduction, Accumulation, and Technical Change 70

Crisis Theory 74

Bibliography 78

vi

5. Exploitation 79

Introduction79

Exploitation, Freedom, and Force81

Exploitation in History84

Exploitation and Justice92

Bibliography101

6. Historical Materialism103

Introduction103

The Development of the Productive Forces105

Base and Superstructure112

The Stages of Historical Development117

Bibliography121

7. Class Consciousness and Class Struggle122

Introduction122

The Concept of Class123

Class Consciousness129

Class Struggle134

Bibliography139

8. Marx's Theory of Politics141

Introduction141

The Capitalist State143

Politics in the Transition to Capitalism153

Politics in the Transition to Communism159

Bibliography166

9. The Marxist Critique of Ideology168

Introduction168

Political Ideologies173

Economic Thought as Ideology176

Religion as Ideology180

Bibliography184

10. What Is Living and What Is Dead in the

Philosophy of Marx?186

Introduction186

What Is Dead?188

What Is Living?194

vii

PREFACE

In 1985 I published a lengthy book on Marx, Making Sense of Marx (Cambridge University Press). The present book is much shorter, about 25 percent of the

first. It has virtually no exegetical discussions of the texts or of the views of other Marxist scholars. The main intention is simply to state Marx's

views and engage in an argument with them. With two exceptions, there is little here that is not found, in some place and some form, in the first book.

In Chapter 1, I provide a brief bio-bibliographic survey that is not included in Making Sense of Marx. In Chapter 3, I offered discussion of alienation

that goes substantially beyond what was found in the earlier book. A fuller development of the ideas sketched there is found in my "Self-realization in

work and politics," Social Philosophy and Policy (1986).

A companion volume of selected texts by Marx, organized along thematic lines corresponding to Chapters 2-9, is published simultaneously with this book.

viii1

1 OVERVIEW

INTRODUCTION

ONE hundred years after his death, Marx is an enormous presence among us. On purely quantitative criteria, judged by the number of his self-avowed followers,

he exerts a greater influence than any of the religious founders or any other political figures. His doctrine being secular rather than timeless, we would

not expect it to have the staying power of Islam, Christianity, or Buddhism, but up to now it has shown few signs of waning. It is not difficult to justify

a continued interest in his writings.

The interest may be extrinsic or intrinsic. One may go to Marx to understand the regimes that have been influenced by him, or to understand and assess his

writings as if he had had no posterity whatever. Of these, the former requires the latter, but not the other way around. When a doctrine - be it religious

or political - becomes an institutional force, it always becomes the object of intense scrutiny in its own right, because the proper interpretation may

be a matter of momentous importance. This is not to say that all dogmatic controversies are decided on purely internal criteria of validity or consistency.

Many of them owe their resolution to mundane struggles of power, in which, however, purely textual arguments serve as one form of ammunition. Although

textual considerations and rational assessment by themselves probably do not set constraints on the outcome, they may in some cases tip the balance one

way or the other. The student of political processes in contemporary communist societies will do well, therefore, to know the texts that form part of the

arsenal of debate. Although Engels and Lenin are the more frequently cited, Marx provides the final touchstone.

2

The guiding interest in the present exposition is, however, purely intrinsic. This will be taken to mean three things. First, it is a matter of establishing

what Marx thought. This task is subject to the usual principles of textual analysis: to understand each part in the light of the whole and when in doubt

to choose the reading that makes the texts appear as plausible and as consistent as possible. In Marx's case the task presents unusual difficulties. For

one thing, the bulk of the corpus consists of unpublished manuscripts and letters in a very uneven state of completion. Some of them, though preserved

for posterity, are still unpublished, so that no interpretation can claim to be based on all surviving texts. For another, the published writings are largely

journalistic or a gitatorial and as such are unreliable guides to his considered opinion. There is, moreover, the problem of ascertaining what was written

by Marx and what by Engels and whether the latter's writings can be used as evidence for Marx's views. Finally, one must take account of the fact that

Marx's thought changed over time, including both discontinuous breaks and more gradual evolution.

We have, in fact, only two published- writings that show us Marx at the height of his theoretical powers: The Eighteenth Brumaire of Louis Bonaparte and

the first volume of Capital. They make up approximately one thousand pages, of a corpus of perhaps thirty thousand. They form, as it were, the fixed point

from which the other writings may be surveyed and guide the choice between different readings. They do not suffice, however, to eliminate all ambiguities

- among other things because they are far from perfectly clear and consistent themselves. Even in his most carefully written works, Marx's intellectual

energy was not matched by a comparable level of intellectual discipline. His intellectual profile is a complex blend of relentless search for truth, wishful

thinking, and polemical intent. Between the reality he observed and his writings, there intervened at least two distorting prisms, first in the formation

of his thought and then in the way he chose to express it.

The operation of the first kind of bias is most evident in his views on communist society - whether communism as he conceived it was at all possible, and

whether it would come about in the course of history. He seems to have proceeded on two implicit assumptions:

3

First, whatever is desirable is feasible; second, whatever is desirable and feasible is inevitable. The second kind of bias is most clearly seen in his

political writings. There is the bias of compromise, due to the need to reconcile different factions; the bias of exhortation, arising from the desire

to use the analysis of the situation" as a means to changing it; and the bias of anticipated censorship, which operated when he had to disguise or tone

down his views to be allowed to state them at all.

Next, I shall ask whether Marx was right in what he thought on the numerous issues - historical as well as theoretical - that he confronted. This examination

will involve deliberate anachronism in the sense that it will draw on facts and theories not available to Marx. In particular, the exposition of Marx's

economic theories will use language developed much after his death. I shall also have the occasion to point out that on various factual matters Marx has

been proved wrong by more recent scholarship. In fact, by and large it will appear that strictly speaking Marx was almost never "right." His facts were

defective by the standards of modern scholarship his reckless and sweeping. ,

A more interesting question, however, is whether Marx remains ' useful for us today. Which of Marx's theories are hopelessly dated or dead, and which remain

a source of new ideas and hypotheses? To answer this question we must look at the wood, not just at the trees. As in the somewhat similar case of Freud,

we may find that a theory can be shot through with errors of detail, even have basic conceptual flaws, yet remain immensely fertile in its overall conception.

It is in the nature of the case that such assessments must be somewhat vague. The Marxian ancestry of a given line of inquiry may not be obvious and is

certainly not proven by the claim of its practitioners to be among his descendants. Yet there exist un-mistakably Marxian theories of alienation, exploitation,

technical change, class struggle, and ideology that remain viable and vital

The organizing idea of the exposition, therefore, is to set out what I believe were Marx's views on the central issues before him, to assess their validity

in the light of the best knowledge available to us today, and to discuss whether the general conceptions underlying them can be useful even when his specific

implementation is flawed. The range of issues covers normative as well as explanatory

4

problems. I take the view that Marxism includes both a specific conception of the good life and a specific notion of distributive justice, in addition to

a theory of history and an analysis of capitalism. The emphasis on normative issues is probably the most distinctive and controversial feature of the exposition.

Most other commentators affirm that Marx denied the existence of absolute values, some of them seeing in this a cause for praise and others for blame.

Can one be a Marxist today? The overriding goal of the exposition is to help the reader form an answer to this question. Many would say, both on intellectual

and on moral grounds, that it is no longer possible to be a "Marxist." Many of Marx's most cherished doctrines have been totally demolished by argument.

Others have been refuted by history, which has shown us that the logical consequence of his political philosophy is an abhorrent social system. What little

remains can be and largely has been absorbed into mainstream social thought.

Each of these three arguments may be countered. To the question whether I am a Marxist, or why, on intellectual grounds, I would want to call myself a Marxist,

I have a well-rehearsed answer: "If, by a Marxist, you mean someone who holds all the beliefs that Marx himself thought were his most important ideas,

including scientific socialism, the labor theory of value, the theory of the falling rate of profit, the unity of theory and practice in revolutionary

struggle, and the Utopian vision of a transparent communist society unconstrained by scarcity, then I am certainly not a Marxist. But if, by a Marxist,

you mean someone who can trace the ancestry of his most important beliefs back to Marx, then I am indeed a Marxist. For me this includes, notably, the

dialectical method and the theory of alienation, exploitation, and class struggle, in a suitably revised and-generalized form."

Among intellectuals in Eastern Europe, with few exceptions, "Marxism" is a dirty word. To them it signifies not the liberation but the oppression of man.

The view is encapsulated in Solzhenitsyn's refusal to meet Sartre in Moscow and memorably argued in Kolakowski's Main Currents of Marxism. It is an attitude

that commands great respect, but its implications for the understanding of Marx are somewhat unclear. True, the work of Marx was one of

5

the causes that led to the Soviet regime; equally true, that regime justifies itself through Marx, asserting that it is roughly the kind of regime he wanted

to bring about. That assertion is manifestly false. Yet the real question lies elsewhere. It is whether any attempt to bring about the kind of regime he

wanted necessarily has to employ means that will in fact bring about something roughly similar to the Soviet regime. This I deny. Yet I shall also argue

that an attempt to achieve the goal by means of a violent proletarian revolution will be self-defeating. The revolutionary bid for power can succeed only

under conditions of backwardness that will also prevent, not only initially but indefinitely, the flowering of the productive forces that Marx posed as

a condition for communism as he understood it.

It would seem, finally, that Marxism as a body of positive social theory, concerned with establishing and explaining facts, ought to disappear if it is

bad and also if it is any good. In the latter case its findings will enter the main body of the historical and social sciences and cease to be specifically

"Marxist." The identity and survival of Marxism is linked, however, to its normative foundation. Because of their adherence to specific, not universally

shared values, Marxist scholars ask different questions. In arguing for their answers, on the other hand, they have to follow the same canons of method

and reasoning as other scholars. Because of their values they look for different things to explain, but the logic of explanation is the same. Their theories

will, if plausible, enter the mainstream of social science if they can also be useful to scholars who ask other questions; and if they cannot it is a good

bet that they are not very plausible.

MARX: LIFE AND WRITINGS

Because of the great variety and diversity of Marx's writings, it is often useful to know when, under which circumstances, for which purposes, and for which

public they were written. The following chronological survey of his writings is meant to facilitate the more systematic discussions in later chapters.

It is not intended as a biographical sketch. Only information about Marx's life directly relevant to the understanding of his work is included.

6

1818-1835: Trier. Marx grew up in the town of Trier in Prussian Rhineland, a formerly liberal province now under a harshly oppressive regime. Both his parents

descended from rabbinical families, but his father converted the whole family to Protestantism to escape discrimination against Jews. Much has been made

of Marx's Jewish background and the alleged self-hatred that led him to espouse anti-Semitism. There is something to the allegation, but Marx's anti-Semitism

never took a virulent practical form. His attitude toward the Slavonic peoples - his "Russophobia" - was in fact more deeply shaped by racism.

1835-1841: University studies. Upon leaving school, Marx studied briefly at the University of Bonn and then for five years in Berlin. Here he came to know

the philosophy of Hegel and met a group of left-wing philosophers known as the "Young Hegelians," who were mainly concerned with the critique of religion.

He wrote his doctoral dissertation on "The Difference between the Philosophies of Nature in Democritus and Epicurus," an echo of which is found in the

frequent references in later works to the trading nations who live "in the pores of society, like the Gods of Epicurus." The Hegelian imprint these years

gave to his thinking never wore off completely, although it is not equally apparent in all his writings.

1842-1843: Journalism and philosophy. During this period Marx worked as journalist and then as editor for the Cologne newspaper Rheinische Zeitung. His

articles show him to be a radical liberal, concerned with freedom of the press and protection of the poor, without, however, seeing the latter as the agent

of their own emancipation. After the paper was suppressed by the government in early 1843, Marx devoted a long summer to philosophical studies. One fruit

of this activity is The Critique of Hegel's Philosophy of Right, a commentary on 261-313 of Hegel's work. The work was first published in this century.

Another, the essay "On the Jewish Question," shows Marx from his worst side as a writer. It is replete with overblown and obscure rhetoric as well as offending

remarks about Judaism. It remains of some interest, however, because it contains Marx's only statement on the rights of man, which he characterizes as

"the rights of egoistic man, of man separated from other men and from the community."

1843-1845: Paris and communism. From late 1843 to early 1845

7

Marx lived in Paris. He became a communist and in the article "Critique of Hegel's Philosophy of Right: Introduction" stated his belief that the proletariat

must emancipate itself and thereby the whole of society. He also wrote a long critique of capitalism, variously known as The Paris Manuscripts or Economic

and Philosophical Manuscripts of 1844, published only in this century. The notion of the alienation of man under capitalism is the central theme. In paris

he also began his lifelong friendship with Friedrich Engels and collaborated with him on an exuberantly juvenile refutation of the Young Hegelians, The

Holy Family; or, Critique of Critical Criticism

1845-1848: Brussels. Marx had been active among the emigre German politicians in Paris and, as a result of pressure from the Prussian government, had to

leave Paris for Brussels. He remained active in politics, first on a local and then on a European scale. Three important writings punctuate these years.

In 1845-6 he and Engels collaborated on the posthumously published German Ideology, in which historical materialism emerged in full-fledged form or at

least as fully fledged as it ever came to be. In 1847 he published Misere de la Philosophie, a reply to P.-J. Proudhon's Philosophie de la Misere, which

had appeared the previous year. In the heavily ironic style he had not yet discarded, Marx makes fun of Proudhon's attempt to master the Hegelian dialectic

and of his petty-bourgeois outlook. In 1848 Marx and Engels collaborated on The Communist Manifesto, published in London by the Communist League. This

masterpiece of political propaganda contains a sweeping historical overview and extravagant praise of the civilizing power of capitalism, concluding that

"what the bourgeoisie produces, above all, is its own gravediggers." By this time two of the three major pieces of Marx's doctrine were in the theory of

alienation and historical materialism. The theory of exploitation existed in an embryonic stage but was not fully worked out until many years later.

1848-1849: Revolution in Germany. On 26 February 1848 news of the revolution in Paris reached Brussels. A week later Marx arrived in Paris and left for

Cologne in early April to become editor of the Neue Rheinische Zeitung, which published 300 issues before it folded in May 1849. In its pages Marx initially

encouraged the

8

German bourgeoisie to pursue the work of the democratic revolution, but when they shied away from what in his view was their historical mission, his policy

took a leftward turn. He could not, however, stem the counterrevolutionary tide. When he was expelled from Germany in May 1849, he also left active politics

for fifteen years, not counting emigre squabbles in London.

1850-1852: The sociology of French politics. From August 1849 to his death Marx lived in London, interrupted by a few brief visits abroad. In the short-lived

Neue Rheinische Zeitung: Politis chokonomische Revue, he wrote a series of articles on French politics, which were published by Engels in 1895 as The Class

Struggles in France. They cover the period from the outbreak of the February Revolution to August 1850. The Eighteenth Brumaire of Louis Bonaparte was

published in 1852 and covers the whole period from 1848 to Louis Napoleon's coup d'etat in December 1851. These writings remain our main source for the

understanding of Marx's theory of the capitalist state, together with the contemporary articles on English politics.

1850-1878: Economic studies and writings. In June 1850 Marx obtained a ticket to the Reading Room in the British Museum, thus beginning the economic studies

that eventually led to the three volumes of Capital. The road was long, twisted, and thorny. Of the numerous manuscripts Marx produced in these years,

only two - Critique of Political Economy and Capital I - were published in his lifetime. The publication of the others has been scattered over a century,

from 1884 to 1982. To help the reader orient himself in this wilderness, I shall indicate the date of writing and of publication of these manuscripts,

as well as their relation to one another.

1857-1858: Grundrisse der Kritik der politischen Okonomie. This huge manuscript - 1,000 printed pages - was first published in Moscow in 1939-41 but was

not available to Western scholars until the East German publication in 1953. It is partly an impenetrable Hegelian thicket, partly a wonderfully inspiring

study of economic philosophy and economic history. It is perhaps the freshest and most engaging of all Marx's works, one in which his ideas can be studied

in statu nascendi.

1859: A Critique of Political Economy. This work corresponds to the first and least interesting part of the Grundrisse. Its place in the

9

history of Marxism is due to the important preface where Marx summarized, in a single long paragraph, the basic tenets of historical materialism. Until

the publication in 1926 of the first part of The German Ideology, these few sentences remained the only authoritative statement of the theory.

1861-1863: Zur Kritik der politischen Okonomie. This enormous manuscript - 2,300 printed pages - consists of twenty-one notebooks. Of these, notebooks 6

through 15 were published by Karl Kautsky in 1906-8 as Theories of Surplus-Value. They deal mainly with the history of economic thought, but contain also

important substantive passages. The remaining notebooks were published in 1976-82, in the new scholarly edition of Marx's and Engels's collected works

(see the section, "Editions of Marx's Writings"). They are preliminary studies to the first and third volumes of Capital, and supplement them on some points.

1865: Results of the Immediate Process of Production. This manuscript was published in Moscow in 1933 but was first made available to Western scholars in

1969. It was intended to serve as a bridge between the first and the second volumes of Capital.

1867: Capital I. This book is beyond doubt and comparison Marx's most important work. It stands with Darwin's Origin of Species as the most influential

book of the nineteenth century. Although Marx intended it to serve the cause of the working class, it is also and preeminently a book for the happy few,

by one of them. Marx assumes that his readers know Greek, Latin, and the major European languages; that they are capable of recognizing remote allusions

to literary and philosophical works, besides being thoroughly familiar with arcane matters of political economy. It is carried by a white-burning indignation

that is all the more effective for being more disciplined than in Marx's earlier works, which were often marred by heavy sarcasm. Considered as economic

analysis it was not a lasting achievement, but it remains unsurpassed as a study of technical change, entrepreneurial behavior, and class conflict in the

age of classical capitalism.

1865-1878: Capital II. This work, published by Engels in 1884, is with a few exceptions utterly boring and, unlike the two other volumes, does not repay

reading for anyone but Marx scholars. Marx's attempt to draw interesting conclusions from simple accounting

10

principles was not successful. The exceptions are the schemes of simple and extended reproduction, which anticipate modern input-output analysis, although

the gist of Marx's analysis can be stated in one page whereas he uses more than one hundred.

1864-1875: Capital III. This work, published by Engels in 1894, is much more valuable. The chapters on economic history are among Marx's most important

writings. The chapters on value theory and crisis theory contain his most authoritative statements on these topics. It must be added, however, that the

scholarly consensus today is that these theories are seriously, perhaps irreparably, flawed. The non specialist reader will not profit much from struggling

with Marx's exposition of them.

1852-1862: American journalism. To earn a living, Marx wrote about five hundred articles, over a period of ten years, as European correspondent for the

New York Daily Tribune, a leading American newspaper of progressive persuasions. Many of his contributions are small masterpieces of historical and political

analysis, notably the articles on the British rule in India and the numerous articles on English politics, which form a useful supplement to the writings

on France. In others one is more struck by his bias than his acumen, as when he touches upon one of his betes noires, the British foreign minister Lord

Palmerston or Napoleon III.

1864-1872: The First International. Marx played a leading, in fact dominating, role in the International Working Men's Association, an organization of European

trade unions. Marx penned the inaugural address and the provisional rules of the International and was elected to the General Council, which was in charge

of day-to-day affairs between the annual congresses. Its first years were marked by Marx's successful struggle against one anarchist faction, the followers

of Proudhon; the last years by the unsuccessful struggle against another, grouped around Mikhail Bakunin. The most important written work from this period

is The Civil War in France, a postmortem on the revolutionary insurrection in 1871 known as the Paris Commune.

1873-1883: Last years. The last years of Marx's life were marked by ill health. He worked on the manuscripts for Capital but without much progress. He guided

from a distance the emerging working-

11

class movement in Germany and wrote an important commentary - The Critique of the Gotha Program - on a document that was drawn up when the two socialist

parties in Germany merged in 1875. He took an interest in Russian history and society and corresponded with Russian socialists about the proper strategy

for revolution in a backward country not yet permeated by capitalism.

MARX AND ENGELS

Friedrich Engels (1820-95) collaborated closely with Marx over a period of forty years. In the eyes of posterity, especially in the communist countries,

they have merged into one entity, Marx-and-Engels. Even scholars sometimes assume without much argument that statements by Engels can be used to support

this or that interpretation of Marx. The present exposition is guided by the opposite principle: Only statements by Marx are used to argue that Marx held

this or that view.

Marx was a genius, a force of nature. Engels was a minor, prolific, somewhat pedantic writer. He began the tradition of codifying Marx's thought into a

total system that promises answers to all questions in philosophy, the natural sciences, and the social sciences. The polemical work Anti-Duhring, in particular,

became immensely influential, especially the discussion of dialectics. It is a fact of major tragicomical proportions that a third of mankind professes

these naive, amateurish speculations as its official philosophy. It is often argued that because Marx read the manuscript to Anti-Duhring, and as far as

we know did not object to it, its views can be imputed to him en bloc. The argument is worthless. Marx was constitutionally incapable of arriving at his

conclusions without deep, prolonged, and independent study - always seeking out the original sources and developing his own views only when he had thoroughly

assimilated them. It is an attitude totally foreign to secondhand acceptance of ideas. He had the best of personal reasons for taking an interest in Engels's

work and occasionally referring to it, but this does not warrant the view that he fully endorsed it. There are, of course, even stronger objections to

imputing to him the views on historical materialism voiced by Engels after Marx's death.

12

It seems justified, however, to use the works written jointly by Marx and Engels - above all, The German Ideology and The Communist Manifesto - as evidence

for Marx's views. If we compare the powerful, complex arguments of the former work with Engels's contemporary writings on similar topics, it is difficult

to believe that he had more than a small share of the responsibility. Similarly, if we compare the latter work with the rough draft, written by Engels

alone, we must conclude that the broad historical sweep and the most penetrating formulations are due to Marx. In any case - and this is what matters -

there is no reason to believe that Marx did not identify himself fully with the views expressed in the two works.

MARXISM AFTER MARX

The development of Marx's doctrine after his death first followed the course of a mainstream, the Second International, and then divided into two separate

currents Soviet Marxism and Western Marxism The story of these developments is, by and large, a depressing one. Although the Marxist movement has produced

some great political leaders, there have been no outstanding thinkers after Marx. Moreover, the propensity of some political leaders to believe themselves

great thinkers and their ability to impose this view on others have had a permanently stultifying effect on intellectual life in the communist countries.

Marxists in Western Europe, on the other hand have indulged in obscurantism, utopianism, and irresponsibility. There are nuances and exceptions, but on

the whole it is difficult not to. subscribe to Kolakowski's negative assessment of the development of Marxist doctrine.

The Second International was formed in 1889 as an association of (mainly European) socialist parties. For all practical purposes it broke down in 1914,

when workers of different countries took to arms against each other. Whatever was left was destroyed a few years later, when the October Revolution made

it evident that the carefully worked out compromise formulations did not provide any guide to hard political choices.

Politically as well as theoretically, the International was dominated by the German Socialist Party (SPD). Although its official.

13

image was that of the revolutionary spearhead of the working class, it was in reality a conservative, bureaucratic organization, oriented mainly toward

its own survival and entrenchment. The sociologist Robert Michels cited the SPD as evidence for what he baptized the "iron law of oligarchy." A later historian

referred to its strategy as "negative integration" and "revolutionary waiting." The leading theoretician of SPD, Karl Kautsky, was also the dominant thinker

of the International, together with the Russian Georghi Plekhanov. Between them they completed the process begun by Engels - to cut the brilliant, sometimes

incoherent ideas of Marx down to size and order. With some finishing touches added by Lenin, "Marxism-Leninism" with the twin doctrines of his, torical

materialism and dialectical materialism - was in place. It is characterized by shallow Hegelianism, naive scientism, lack of falsifiability, and a strong

preference for assertion over argument. It is Marxism set in concrete.

There were other trends and figures in the International. An early revolt against the pseudo revolutionary stance of the SPD was made by Eduard Bernstein

around 1900. He asserted, essentially, that revolution was unlikely, because capitalism was no longer prone to cyclical crises; superfluous, because the

socialist goals could be realized by nonviolent means; and in any case undesirable, because notions like "the dictatorship of the proletariat" belong to

a lower stage of civilization. Although these views largely coincided with the practices of SPD, the party was embarrassed by his voicing them in public.

"Dear Ede," wrote a trade union leader, "one doesn't say these things, one simply does them."

Left-wing critics, notably Lenin and Rosa Luxemburg, wanted the International to become genuinely revolutionary. The "organizational question" is central

in their writings. Marx had never developed a theory of organization, except for the general remark that the emancipation of the working classes must be

conquered by the working classes themselves," because the opposite view runs into the problem that "the educator must himself be educated." To Lenin's

mind, this attitude was sheer romanticism. with relentless pragmatism, he insisted on a centralized and hierarchical organization of the workers - what

came to be known as

14

"democratic centralism." Rosa Luxemburg, on the other hand, tried to work out the implications of Marx's views, toward a working-class movement that was

both spontaneous and revolutionary. She was the first great "activist" of the socialist movement - precursor and heroine of May 1968 but with a more serious

bent than the flower generation. She is the only one of the great socialist leaders in the West to have been killed in revolutionary action, in Berlin

after the end of World War I.

Luxemburg was also one of the best analytical minds of the Second International. The crispness and freshness of her political writings strike one even today,

although her more self-conscious theoretical efforts are distinctly forgettable. An even more impressive thinker was Leon Trotsky Like Marx and like Luxemburg,

he suffered from wishful thinking and lack of intellectual discipline, but he also had a rare grasp of history and political sociology, which enabled him

to adapt Marx's theory of revolution to backward nations. It can be argued, however, that history has shown him right to a greater extent than he hoped

for, by suggesting that revolutions will occur only in the countries on the periphery of capitalism, without spreading to the core countries.

The further history of Marxism in the communist countries resembles that of any other "degenerating research programme," to use a phrase from Imre Lakatos.

The development of the theory took the form of Ptolemaic additions to save appearances, ad hoc hypotheses to explain anomalies, tortuous exegeses, and

stubborn disregard of facts. It was accompanied by total destruction of philosophy, with the partial exception of logical theory; by near-total paralysis

of the social sciences; and by a severe setback for the natural sciences, notably in genetics. The destruction of reason that took place under Stalin or

during the Chinese Cultural Revolution had no precedent in history. The recovery is still incomplete or uncertain, except in the natural sciences.

The causal role of Marxism-Leninism in these developments remains unclear. The rise of the pseudo geneticist Lysenko was probably due more to his proletarian

background and to the immense faith in the power of science to bring rapid results that characterized the Soviet state in the first years than to any "dialectical"

features of his views. Indeed, by virtue of its essential

15

vagueness dialectical materialism lends itself to the justification, after the fact, of virtually any theory. Whereas some argued that dialectical materialism

requires genes to have a specific material substrate, others asserted that dialectical materialism, as opposed to mechanical materialism, requires them

to be lodged in the organism as a whole. Whether the choice between these views was decided on scientific grounds or on political, the philosophical justification

was mainly a ritual embellishment.

Most forms of Western Marxism can be characterized as attempts to create a synthesis of Marx and various other thinkers. Its inception was marked by the

publication of Georg Lukacs's History '. and Class Consciousness in 1923. This work anticipates in a quite, remarkable way the Hegelian reinterpretation

of Marx that was fully launched a few years later, with the publication of Marx's early manuscripts. Lukacs was also influenced by the writings of Max

Weber; in particular, his interpretation of Marx's notion of alienation owes much to Weber's idea of the increasing rationalization of society. In spite

of many valuable insights in this and other works, Lukacs's career as a whole can be summarized as the abdication of reason or, in Kolakowski's phrase,

reason in the service of dogma. The political and intellectual irresponsibility of his work is matched only by the fascination that for a long time he

exerted on other Western intellectuals. Like Pascal with regard to religion, they argued that in politics "il n'y a rien de si conforme a la raison que

ce desaveu de la raison."

Within the same intellectual orbit we may also mention figures like Karl Mannheim or Karl Korsch. More durably influential, however, was the "Frankfurt

School." Counted among its original members were Herbert Marcuse Max Horkheimer. and Theodor von Adorno; the central contemporary descendant is Jurgen

Habermas. Theirs was a synthesis not of Marx and Weber but of Marx and Freud - two great unmaskers debunkers. and would be liberators. Much of the work

of the early Frankfurt School is marred, however, by Hegelian obscurantism and thinly disguised elitism. Again the reader is referred to Kolakowski for

a devastating review. The work of Habermas is also somewhat impenetrable but more solidly founded in rational argument. French Marxism has been through

two major phases. The first

16

was dominated by Jean-Paul Sartre and Maurice Merleau-Ponty, who added Husserl and Heidegger to the prisms through which Marx could be read. It is somewhat

inaccurate, however, to refer to this school as "existential Marxism," because the major work it produced - Sartre's Critique of Dialectical Reason - owes

less to existential philosophy than to French economic and political historians.The second phase arose through the improbable and barren marriage of Marx

with Ferdinand de Saussure, the founder of structural linguistics. In the interpretation of Louis Althusser, scientism again came to the forefront in Marxism,

after a generation of Marxists who had insisted that the categories of natural science were inapplicable to the study of society.

English Marxism: Is there such a thing? Marx himself grew increasingly cross with his adopted country, deploring the crude empiricism of the English and

their lack of revolutionary fervor. The Hegelian aspects of Marxism certainly never took hold in England or in the United States. Marxian political economy,

which did not much interest the Continental Marxists, had stronger appeal. Joan Robinson's Accumulation of Capital from 1956 was perhaps the most important

work in Marxist economic theory after Marx, although it turned out to create an orthodoxy of its own that has become a serious obstacle to progress. Other

recent Anglo-American work in Marxist philosophy, history, and social science is more promising - at least in my opinion, because I am referring to the

tendency that has shaped the present exposition. The work of GA Cohen, John Roemer and others unites rigor and relevance in a way that has been Sorely

lacking in Marxism.

There is, finally, the Marxism of the Third World countries. This has largely been concerned with extending Marx's notion of exploitation from the national

to the international domain. Examples include the dependency theory of Andre Gunder Frank, the theory of unequal exchange proposed by Aghiri Emmanuel,

and Samir Amin's theory of accumulation on a world scale. Though often suggestive, these writings are with few exceptions flawed by technical deficiencies

and conceptual naivete.

Because these writers do not in these respects compare unfavorably with, say, Horkheimer or Althusser, this comment ought

17

not to be taken as a slur on their Third World origin. The problem is that Marxism tends to attract thinkers who are either fascinated by the darkly Hegelian

origins of the theory or - on the other extreme - urgently preoccupied with practical relevance. Often the extremes meet, in an unnerving combination of

extremely abstract theory and highly specific proposals. What Marxism needs is the development of what Robert Merton called "theories of the middle range."

For this purpose it may be necessary - indeed, in the present phase it is necessary - to think less about relevance in the short term, to become more relevant

in the long term. When Marx went into his inner exile in the British Museum, he followed the strategy "One step backward, two steps forward," taking time

off from politics to fashion a tool that could then be of use in politics. The theory he developed has done service for a century but is becoming increasingly

irrelevant for most of our urgent problems. "Back to the British Museum!" is hardly a slogan with mass political appeal, but it is one that Marxists would

do well to ponder.

EDITIONS OF MARX'S WRITINGS

In German and English, there are four editions of Marx's and Engels's collected works, none of them complete. There are also numerous editions of individual

writings, a few of which are mentioned below.

There are two different editions, each of them intended as the definitive scholarly edition, referred to as Marx-Engels Gesamt-Ausgabe (MEGA). The first

MEGA was published between 1927 and 1935. Before it was stopped by Stalin's rise to power, it had transformed Marx scholarship and Marxist thought through

the publication of The Economic and Philosophical Manuscripts and The German Ideology.

The next edition, in chronological order, is the Marx-Engels Werke (MEW), published in East Berlin from 1956. It is the only edition that is approximately

complete, but it is not a scholarly edition. Marx's writings in English and French are published in German translation. Many of the posthumous economic

writings

18

are not included here; some works also appear to have been excluded on ideological grounds. The editorial introductions and notes are heavily dogmatic but

provide much useful information.

An English translation of MEW has been in progress since the 1970s, as Karl Marx and Friedrich Engels, Collected Works (CW). It appears to aim at completeness

but not at scholarship; yet it is invaluable to Marx scholars because much of the English journalism here becomes easily accessible in the original language

for the first time.

Finally, the second MEGA has been unfolding since 1977 at a majestic pace and will not come to a halt until well into the next century. It has already brought

us several major unpublished manuscripts, notably the previously unknown parts of the 1861-3 Critique. More preparatory manuscripts for Capital will follow.

The scholarly apparatus is splendid, but the editors have not felt able to avoid the usual ideological rituals.

The standard German editions of the three volumes of Capital are identical with volumes 23-5 of MEW. An English translation of the first volume appeared

in 1886, authorized by Engels. All later editions have been reprinted of this translation, until a new translation by Ben Fowkes was published by Penguin

in 1976. Being as reliable as the original translation and less encumbered by heavy Victorian prose, it ought eventually to replace it as the standard

English reference. Yet for many years to come many will continue to read the three volumes in the edition published by International Publishers, as this

is the one usually cited by English scholars.

The 1953 German edition of the Grundrisse has now been superseded by the publication in the new MEGA. An English translation by Martin Nicolaus was published

by Penguin in 1973. The standard German edition of the Theories of Surplus-Value used to be volumes 26.1-26.3 of the MEW, but these are also superseded

by the new MEGA. The standard English translation is published by Lawrence and Wishart. The Results of the Immediate Process of Production is translated

as an appendix to the Penguin edition of Capital I; an easily accessible German edition was published by Verlag Neue Kritik (Frankfurt) in 1969, itself

a photographic reprint of the 1933 Moscow edition.

19

There are many books entitled "Marx on ..." or "Marx and Engels on . . . ," collecting their writings and obiter dicta on various topics. There are also

many selections of their writings from different periods: the early manuscripts, the American journalism, writings from the First International, and so

on. There are various "Selected Writings" covering the corpus as a whole and there are, of course, many editions of most individual writings. It is impossible

to list them all, and there is not much point in selecting a few. If one should be singled out, it is Saul Padover's edition of The Letters of Karl Marx,

(Prentice-Hall, 1979), which usefully collects the most important correspondence. Besides providing valuable supplementary sources for the understanding

of his theories, Marx's letters vividly convey his ebullient personality.

BIBLIOGRAPHY

Introduction. By far the best guide to the Marxist universe is Leszek Kolakowski's Main Currents of Marxism, 3 vols. (Oxford University Press, 1978). It

is excellent on philosophy, superb on politics, but somewhat scanty on social theory, especially economics.

Life and writings. There is no first-rate, full-length biography or intellectual biography of Marx comparable to those of Rosa Luxemburg or Leon Trotsky

cited below. An excellent short study is Isaiah Berlin's Karl Marx, 3d ed. (Oxford University Press, 1973). Maximilien Rubel's Karl Marx: Essai de Biographic

Intellectuelle (Presses Universitaires de France, 1959) remains useful. A clear biographical presentation is David McLellan, Karl Marx: His Life and Thought

(Macmillan, 1973). More ambitious, with deeper psychological understanding, is Jerrold Seigel, Marx's Fate: The Shape of a Life (Princeton University Press,

1978). On Marx's Jewishness, see Julius Carlebach, Karl Marx and the Radical Critique of Judaism (Rout-ledge and Kegan Paul, 1978) and Isaiah Berlin, "Benjamin

Disraeli, Karl Marx, and the search for identity," in his Against the Current (Viking Press, 1980). Two books on Marx as a politician are Oscar J. Hammen,

The Red '48ers: Karl Marx and Friedrich Engels (Scribner, 1969), and Henry Collins and Chimen Abramsky, Karl Marx and the British Labour Movement (Macmillan,

1965). Marx's relation to the anarchists is discussed in Pierre Ansart, Marx et l'Anarchisme (Presses Universitaires de France, 1969), and in Paul Thomas,

Karl Marx and the Anarchists (Routledge and Kegan Paul, 1980). An encyclopedic survey of Marx's stylistic repertoire is S. S. Prawer, Karl Marx and World

Literature (Oxford University Press, 1978).

20

Marx and Engels. A work especially devoted to the relation between Marx and Engels is Norman Levine, The Tragic Deception: Marx contra Engels (Clio Books,

1975). A briefer treatment is that of Gareth Stedman Jones, "Engels and the history of Marxism," in Eric Hobsbawm (ed.). The History of Marxism (Harvester

Press, 1982), 1:290-326.

Marxism after Marx. The three volumes of Kolakowski's Main Currents of Marxism are indispensable. They can be usefully supplemented by volumes 3 to 5 of

G. D. H. Cole's History of Socialist Thought (Macmillan, 1953-60). A good introduction to Soviet Marxism and its historical sources is Z. A. Jordan, The

Evolution of Dialectical Materialism (Macmillan, 1967). It can be usefully supplemented by Loren Graham, Science and Philosophy in the Soviet Union (Knopf,

1972). A history of the SPD is Dieter Groh, Negative Integration und revolutionarer Attentismus (Suhrkamp, 1973). Valuable studies of individual figures

are J. P. Nettl, Rosa Luxemburg (Oxford University Press, 1966), and B. Knei-Paz, The Social and Political Thought of Leon Trotsky (Oxford University Press,

1977). On the Frankfurt School, see Martin_Jay, The Dialectical Imagination (Little, Brown, 1973). There is, to my knowledge, no good treatment of French

Marxism.

Editions of Marx's writings. A useful survey is that of Eric Hobsbawm, "The fortune of Marx's and Engels' writings," in E. Hobsbawm (ed.), History of Marxism,

1: 327-344.

21

2 MARXIST METHODOLOGY

INTRODUCTION

MANY claims have been made for "the Marxist method." Some of them are justified, others are exaggerated, false, or unintelligible. Although Marx had valuable

methodological insights that are not yet fully exhausted, there is no "dialectical reason" that separates Marxists from ordinary mortals. On first exposure

to Marxist writings, many feel mystified and terrorized by references to the "dialectical unity of opposites," the "revolutionary unity of theory and practice,"

and similar phrases. All too often, such locutions have allowed followers of Marx to get away with murder, sometimes literally so. It is against this background

of extreme self-indulgence that I adopt what may look like an excessively purist viewpoint on methodology. Readers may tolerate suggestive ambiguity in

a writer if on past performance they are willing to give him the benefit of doubt, but Marxism has long since exhausted its credit.

The Marxist methodology that I want emphatically to reject is an amalgam of three elements. The first methodological holism the view that in social life

there exist wholes or collectivities, statements about which cannot be reduced to statements about the member individuals. The second is functional explanation,

the attempt to explain social phenomena in terms of their beneficial consequences for someone or something, even when no intention to bring bout these

consequences has been demonstrated. The third is dialectical deduction, a mode of thinking that is derived from Hegel's Logic and that does not lend itself

to brief summary. Of these, the first two are also found, separately or in combination, in non-Marxist social science. Emile Durkheim, among others

22

insisted that, even were psychology to become a perfect science, there would remain some social facts it could not explain. Robert Merton has advocated

explanation of institutions and behavioral patterns in terms of their "latent functions," that is, consequences that were neither intended by the agents

who produce them nor perceived by those who benefit from them. Social anthropologists have proposed explanations that are both holistic/ and functional.

When Marx employs the same method, he differs not only in the Hegelian element sometimes superimposed on it but also in the nature of the wholes and the

benefits that enter into the explanation. Most importantly, he uses functional explanation not only to account for the stability of societies but also

to demonstrate their inherent tendency to develop toward communism.

Marx was very much a nineteenth-century figure, which is to say that in methodological matters he was a transitional figure. He had liberated himself from

explicit theological assumptions, yet he retained the ideological outlook they had inspired. Like most of his contemporaries he was impressed by the progress

of biology and wrongly thought that the study of society could profit from the study of organisms. (It is only fair to add, however, that he indulged much

less in analogical inference from biology than, say, Auguste Comte or Herbert Spencer.) His scientism - the belief that there exist "laws of motion" for

society that operate with "iron necessity" - rested on a naive extrapolation from the achievements of natural science. We shall find, over and over again,

that dated methodological conceptions coexist, in his work, with strikingly fresh insights.

METHODOLOGICAL INDIVIDUALISM,

Methodological individualism is the view that all institutions, behavioral patterns, and social processes can in principle be explained in terms of individuals

only: their actions, properties, and ; relations. It is a form of reductionism, which is to say that it en--4-jems us to explain complex phenomena in terms

of their simpler components. Reductionism is a central strategy in science. It has -created such disciplines as molecular biology and physical chemistry.

The case of biology is especially interesting, because for a

23

long time it was claimed to be inherently irreducible to physics or chemistry. A similar claim is still made today with respect to the social sciences.

The counterclaim is not that there already exists a social psychology or psychological sociology that has effectuated a complete reduction. Rather, it

is that there is no objection in principle to such a reduction being carried out, even though it may remain impracticable for the foreseeable future. One

may add that with respect to some problems the "search for microfoundations" - another term for methodological individualism - has already yielded important

results. A full reduction is possible in principle; a partial reduction is well under way.

This is not the place to defend the doctrine of methodological individualism, beyond saying that our confidence in and understanding of an explanation is

improved if we open "the black box" and get a look at the nuts and bolts, the cogs and wheels of the mechanism inside it. Some objections may in any case

evaporate if a few things are made clear. First, the doctrine has no implications about the kind of individual-level explanation that is needed to carry

out the reduction. In particular, the assumption that individuals are rational and selfish is not part of the doctrine, although compatible with it. Second,

it does not assume that individuals are "atoms" that have a presocial existence before they come together to form society. Relations between individuals

must be let in on the ground floor of social explanation. Third, it does not extend to what goes on inside peoples' heads. In the phrase "The United States

fears the Soviet Union," the first collective noun is subject to reduction but not the second, because what the individual Americans fear may well be a

nebulous collective entity.

Lastly, adherence to methodological individualism should not blind one to the dangers of premature reductionism. An important example is the following.

To understand successful collective action, such as a strike, we would ideally like to give an account in terms of the motives and beliefs of the participants.

These proximate causes of the behavior are, however, usually very elusive. Often the remote causes are more tractable. We may, for instance, be able to

establish a causal connection between, on the one hand, certain features of the group in question and, on the other hand, its Propensity to strike. Other

things being equal, a strike may be

24

more likely the more similar the background of the members and the more stable the composition of the group. We know that the effect of these remote "macro

variables" must be mediated by their impact on individual motives and beliefs, but we may be unable to specify the mechanism. In such cases the search

for micro foundations may lead to barren speculation. The principle of methodological individualism says only that the search is not inherently sterile,

although, in any given case, it may well be pointless in the present state of knowledge.

In Marx there are two main instances of methodological holism, corresponding to his two central theoretical concerns. In the analysis of capitalism, "Capital"

appears as a collective entity, which cannot be reduced to the several individual firms. Marx believed that capital had a logic of its own, which was somehow

prior in the explanatory order to market behavior and competition. According to methodological individualism, on the other hand, any "laws of motion and

self-regulation" of capitalism must be deduced as theorems from axioms specifying the motives and constraints of firms, workers, and consumers.

In historical materialism, "Humanity" appears as the collective subject whose flowering in communism is the final goal of history. Marx was strongly influenced

by Hegel's tripartite division of history into an initial undifferentiated unity, followed by a phase of conflict and alienation and culminating with a

higher unity that retains the individuality developed in the second phase. In Hegel's and Marx's secular theology, mankind had to alienate itself from

itself in order to regain itself in an enriched form. Exactly how this is mediated by the actions of individuals, motivated by goals of their own, is never

made clear.

The point should not be overemphasized. Marx was not a purely speculative social thinker. Most of the time he was indeed concerned with forging links among

individual motives, individual behavior, and aggregate consequences. On the other hand, one should not go to the other extreme and view his references

to Capital and Mankind as rhetorical devices with no explanatory relevance. Marx does often point to the needs of these collective entities in order to

explain events and institutions that appear as if by magic to fulfill them. His belief in the independent logic of

25

aggregates does sometimes weaken his motivation to study the fine grain of social structure and social change. Speculative elements coexist, often within

the same work, with more soundly based views. The Grundrisse, in particular, shows Marx's mind operating in both registers, with dizzying transitions.

"Individualism" is a term with many connotations. Quite independently of the issues just discussed, one may profess individualism in an ethical or normative

sense. This doctrine, though not in itself a substantive ethical view, imposes constraints on any such view by stipulating that in the final analysis only

individuals are morally relevantSome 'implications of this view are the following. The advancement of knowledge, the creation of great works of art, or

the protection of nature are not independent sources of value - they have value only to the extent that they are valued by individuals. Equality between

the sexes, between classes, or between nations is not a value in itself - it is to be promoted only to the extent that it leads to greater equality between

individuals. It is never justified to ask people to sacrifice themselves for the sake of the fatherland or the proletariat - unless one can show that other

concrete individuals benefit. (And even then, of course, the justification may be dubious.)

Marx was an individualist in this normative sense. He appreciated that class societies in general and capitalism in particular had led to enormous advances

in civilization, as judged by the best achievements in art and science. Yet this process was the self-realization of Man rather than of individual men,

who had, for the most part, lived in misery. Indeed, only by the exploitation of the many could class societies create the free time in which a few could

contribute to the progress of civilization. The attraction of communism in his eyes was to allow the self-realization of each and every individual, not

just of a small elite. As a by-product there will also be an unprecedented flowering of mankind, but this is not, to repeat, in itself a source of value.

MARXISM AND RATIONAL CHOICE

That human behavior can be explained in terms of rational choice is becoming a central, perhaps even dominant view in the social

26

sciences. Broadly speaking, it was born around 1870 when the marginalist revolution in economic theory allowed a precise formulation of the costs and benefits

attached to alternative uses of scarce resources. Capital I was published in 1867, so we would not | expect Marx to have been exposed to these ideas. Yet

we may ask ) whether his theory is consistent with the basic assumptions of the rational-choice approach, or whether - as later followers and opponents

have claimed - the two are mutually incompatible.

The explanation of an action makes appeal to two successive filtering processes. From the whole set of abstractly possible courses of action we first filter

out those that do not satisfy the given logical, physical, economic, or mental constraints. Human beings cannot have their cake and eat it too; they cannot

fly in the air like birds; they cannot spend more than they earn; and they cannot act like lightning-fast calculating machines. Within the remaining set

we appeal to some principle of selection that explains which action is finally realized. Rational-choice theory assumes that people will choose the course

of action they prefer, or think is best. To act rationally is to choose the best action in the feasible set. The full structure of rational-choice theory

is more complex than one would glean from this stark statement. Here are some of the qualifications that need to be added. First, although the standard

theory presents constraints and preferences as independent of each other, either can in fact be shaped by the other. The constraints are shaped by the

preferences if the person decides ahead of time to eliminate certain options from the feasible set, which one might do to avoid temptation or more paradoxically

to improve a bargaining position. Conversely, preferences are shaped by constraints if the person consciously or unconsciously adapts what he wants to

what he can get. Second, it ought to be emphasized that the theory says that the person will choose the action which he thinks is best suited to his purposes,

which is not to say ' that it is the best in a more objective sense! Rational-choice explanation embodies a claim about the relation among action, motives,

and beliefs. It differs in this respect from evolutionary explanation in biology, which asserts that organisms end up having the features that are objectively

best from the point of view of their fitness. Third, it often happens that several people simultaneously try

27

to adjust rationally to each other. This belongs to the realm of game theory, as it has somewhat unfortunately come to be known. A better name would have

been the theory of interdependent decisions. I shall say more about game theory later.

First, however, we may go back to the two filters, to see what alternatives to rational-choice theory might look like. Some would deny that there is more

than one filter operating. They would argue that the constraints tend to be so strong as to eliminate all alternatives but one, so that nothing is left

for the second filter to operate on. This view may be plausible in some cases. For workers in classical capitalism, for instance, the joint effect of the

budget constraint and the calorie-protein constraint may have been to narrow the feasible consumption purchases to a very small set. As a general theory,

however, this "structuralist" view is implausible. It is hard to think of a general mechanism that would shape the constraints so as to leave the agents

with exactly one option. It will not do, for example, to say that the members of the ruling class make it their business to restrict the feasible set of

the exploited class, because this statement assumes that the former class has the very freedom of choice it denies to the latter.1

Another class of alternatives proposes different mechanisms in the second filter. There are two main contenders. On the one hand, there is the sociological

view that men are governed by social norms, roles, habit, or tradition. The view implies that behavior is less sensible to changes in the feasible set

than one would expect on the rational-choice approach. Value is attached to specific forms of behavior as such, not only to their outcomes. On the other

hand, there is the view that people do not choose what is best, only what in some sense is good enough. (By a neologism, this view is known as the "satisfying"

theory of choice, as opposed to the "maximizing" assumption of rational-choice theory.) The argument is that the costs of collecting and evaluating information,

as well as the uncertainty surrounding the value of information, make nonsense of the notion of "optimal behavior."

1 In any case, a rational ruling class would leave the exploited class with some freedom of choice. It would restrict the set of alternatives up to the

point where the element preferred within it by the members of the exploited class is also the one which the ruling class prefers within the unrestricted

set.

28

There is a good deal to each of these views. If in my opinion they have not succeeded in dethroning rational-choice theory from its dominant position, it

is because they lack robustness and predictive power. The facts they invoke in their support are real enough, but as is generally acknowledged by philosophers

of science, "Facts don't kick." To discredit a theory it is not sufficient to adduce facts that count against it; one must also produce another, better

explanation. As long as the sociological theory does not specify the limits within which behavior will remain unaffected by changes in the feasible set,

or the satisficing theory does not come up with an explanation of why people have different ideas of what is good enough, they will not be able to claim

superiority. These theories offer what is sometimes called thick description, not explanation.

There is a further response to rational-choice theory that does not quite coincide with any of these alternatives. It goes as follows: True, action can

be explained in terms of the preferences and beliefs of the actors, but this is not a rock-bottom explanation. Motives and beliefs are not identical across

individuals or stable over time. They are shaped and modified by social forces, including deliberate manipulation. Hence, rational-choice explanation offers

a shallow understanding of behavior and must be supplemented by an account of how preferences and beliefs emerge from within the social structure.

The point might appear to be trivial. It is always possible to search, beyond the cause of the phenomenon, for the cause of the cause. One cannot on each

occasion go back to first causes. Yet, for someone who, like the Marxist, wants to understand long-term historical trends, it points to a real difficulty.

In the long run nothing can be taken for constant or given; everything must be explained from within or "endogenously." It is probably true to say that

this is one of the main unresolved problems of the social sciences. True, we do know something about how beliefs and preferences are shaped by social structure.

In particular, there is good evidence for the view that people adjust their desires or their beliefs so as to reduce "cognitive dissonance," that is, the

mental tension that arises when what one values is also believed to be out of reach. The oppressed often end up accepting their state, because

29

the alternative is too hard to live with. Yet we know little about the limits within which this mechanism operates and beyond which revolt becomes a real

possibility.

I promised to say a few words about game theory. I am among those who believe game theory offers a unifying conceptual framework for most of social science,

in that it enables us to understand three kinds of interdependencies that pervade social life. There is, first, the fact that the reward of each depends

on the rewards of all, by altruism, envy, and the like; second the fact that the reward of each depends on the choices of all, through general social causality;

and, third, the fact that the choice of each depends on the choices of all, through anticipation and strategic calculation. This is not to say that we

can observe each kind of interdependency in each case of social action. They represent conceptual possibilities, not necessities.

Game theory has analyzed numerous forms of social interaction. Here I shall only spell out the structure of the most prominent among them, the so-called

Prisoner's Dilemma which is one of the most intensively studied problems in contemporary social science. Rather than retell the anecdote that lent its

name to the dilemma, I shall illustrate it by sticking to the example of strikes. Let us assume that each worker has the choice between two options: to

join the strike or to abstain. Let us assume, moreover, that it is better for all workers if all strike than if none do, because in the former case they

can successfully press for a wage raise. Let us assume, finally, that the workers are solely motivated by personal material gains. Then, for each worker

it is always better to abstain, regardless of what others do. If they strike, he or she can act as a free rider and get the wage rise without the risk

and cost involved in striking. If they do not, there is no point in a unilateral act of solidarity. Hence the individual will not strike; nor, because

they are similarly placed, will the others. The result of all workers acting in an individually rational way is that the outcome is worse for all than

it could have been had they been able to cooperate. This is also often referred to as the free-rider problem.

The main source of Marxist resistance to rational-choice theory is the first, "structuralist" objection. Marx often emphasizes that workers and capitalists

are not agents in the full sense of the term:

30

free, active choosers. Rather, they are mere placeholders or, as he put it, "economic character masks," condemned to act out the logic of the capitalist

system. Workers are forced to sell their labor power, and the idea that they have a free choice in the labor market is an ideological construction. As

consumers, their choice between different consumption plans is restricted by low wages. Similarly, capitalists are forced by competition to act as they

do, including the inhumane practices of exploitation. If they tried to behave differently, they would be wiped out.

This argument fails because the notions of choice and force are not incompatible. Consider two situations. In one, workers have two options: barely surviving

as independent peasants and barely surviving as workers. In the other, the first option is the same, but the other is now to work for a wage that allows

a good standard of living. In the latter case, the workers are forced to sell their labor power - not by coercion but by what Marx calls the force of circumstances.

In the former, the workers clearly have a choice between two options. But if they have a choice in the first situation, they must also have one in the

second; the existence of choice cannot be removed by improving one of the options. A somewhat different argument establishes the reality of capitalist

choice in a competitive market. When capitalists, under the pres- sure of competition, survey alternative modes of action and go through extensive calculation

to find the most profitable, they engage in the very paradigm of choice behavior. The fact that they do not survive if they make the wrong choice does

not mean that they do not make choices; on the contrary, choice is presupposed In Marx's economic theory the denial of choice is closely linked with the

labor theory of value. Marx postulated that the economy had a surface structure and a deep structure. The surface structure is that of everyday economic

life, in which the economic agents make rational choices in terms of the market prices of goods. In the deep structure, goods are characterized by their

labor values - the amount of socially necessary labor time required to produce them. The surface structure is merely the working out of the relations defined

by the deep structure, just as the visible appearance of a physical object is a mere consequence of its atomic structure. I shall argue later that this

theory of the relation between values and

31

prices, the essence and the appearance in economic life, is barely intelligible. It certainly misled Marx, in preventing him from appreciating the centrality

of choice and alternatives in economics. Outside economic analysis proper Marx's structuralist method did not affect his concrete investigations. In the

brilliant core chapters on economic sociology in Capital I, Marx fully recognized the subtle interplay among entrepreneurial choice, profit, technology,

and power in the firm. His political sociology likewise was sensitive to rational and strategic thinking on the part of the main actors. I should add,

however, that in both cases there is one obstacle to a purely rational-choice interpretation of these texts. This is his tendency to deploy functional

explanation, in which events and institutions are explained on the grounds that they are better for some agent or agents but not necessarily on the grounds

that they are chosen because they are better.

FUNCTIONAL EXPLANATION IN MARXISM

Functional explanation is puzzling and controversial, for reasons that may be brought out by comparing it with other modes of scientific explanation. In

causal explanation, we account for a phenomenon by citing its (actual) cause, assumed to have preceded it in time. In intentional explanation, of which

rational-choice explanation is the most important variety, we cite the intended consequences of the phenomenon. Again the intention occurs at an earlier

time than the thing we want to explain by citing it. In functional explanation, we cite the actual consequence of the phenomenon in order to account for

it. Marx, for instance, explains upward social mobility by pointing to the economic benefits the capitalist class derives from having a steady stream of

fresh recruits. The puzzle is how an event can be explained by another event that occurs at a later time. There must be an explanation for the event when

it happens - it cannot be necessary to await the consequences in order to explain it.

The most plausible solution to this puzzle is to deny that functional explanation can account for single events or processes. For something to be a proper

object of functional explanation, it must be a pattern of similar, recurring events. Let me give an example,

32

cited only for the sake of illustration and not because I believe the story it tells is true. If in a given capitalist society we observe upward social

mobility as a regular phenomenon, so that in each generation some workers become self-employed or small capitalists, this pattern could be explained by

the benefits provided to the capitalist class, in the following way. Upward mobility in one generation contributes to the economic vitality and prosperity

of capitalism. A prosperous system provides further opportunities for mobility in the next generation. Moreover, that capitalism is seen to deliver the

goods provides it with legitimacy and channels individual aspirations into desires for mobility within the system rather than revolt against it. The opportunity

and the desire for mobility come together in creating, or re-creating, actual mobility. Hence, upward mobility at one point in time has consequences that

lead to the continued presence of mobility at a later time.

The example shows that functional explanation involves a feedback loop, a causal connection from the consequences of one event of the kind we are trying

to explain to another, later event of the same kind. Functional explanation is applicable when a pattern of behavior maintains itself through the consequences

it, generates; more specifically, through consequences that benefit some group, which may or may not be the same as the group of people displaying the

behavior. In the mobility example the agents and those whom they benefit are distinct groups. An example (again used only for the sake of illustration)

in which they coincide follows.

If the satisficing theory of choice is correct, firms do not and cannot consciously maximize profits. Rather, they make decisions by following rules of

thumb that appear to be "good enough." In a competitive market, however, only the firms that happen to hit upon profit-maximizing rules of thumb will survive;

the others will go bankrupt. We can then explain the observed behavior of firms by pointing to the beneficial consequences of their decision rules. The

reason we observe these particular rules of thumb rather than others is that they maximize the profit of the firm. The pattern of explanation is similar

to that of functional explanation in biology, in which we explain the optimal adaptation of organisms by appealing to chance variation and natural selection.

These two examples provide perfectly valid explanations, assuming

33

the truth of their premises. Why, then, object so strongly to the use of functional explanation in Marxism? First, in Marx's philosophy of history we find

explanations of singular, non recurrent events in terms of their unintended consequences. An argument of this type rests on a metaphysical impossibility.

Second, in many functional explanations - within and outside Marxism - the feedback loop is not demonstrated but only postulated or tacitly assumed. This,

in fact, is the major objection. In the mobility example, the suggested feedback loop is not proposed by Marx, nor does he suggest any other mechanism

that could support the explanation.

The mere fact that an activity has beneficial consequences - be it for capitalist domination, for social integration, or in some other respect - is not

sufficient to explain it. Any phenomenon can be shown to benefit a number of groups or interests, especially if we are allowed to vary the time perspective.

For instance, Marx argued that state policies not in the short-term interest of capitalists might - precisely because of that fact - be in their long-term

interest. If we grant, for the sake of argument, that this is in fact the case, we still have no explanation. This would require the exhibition of a mechanism

by which the satisfaction of long-term interests generates or sustains the policies. In the absence of a demonstrated mechanism, the benefits could, for

all we know, be purely accidental and hence non explanatory.

One important mode of Marxist explanation combines methodological holism and functional explanation by asserting that the behavior of a class, can be explained

by the beneficial consequences for the class members. Consider the following explanation of why technical change tends to be labor-saving rather than capital-saving.

Labor-saving innovations are in the interest of Capital, because they reduce the demand for labor and hence lower the wage that workers have to be paid.

The argument fails because it provides no reason why the individual capitalist firm should prefer this kind of innovation. In a competitive industry, a

single firm is too small to affect the going wage rate and hence has no incentive to search for one kind of innovation rather than another. We are, in

fact, dealing with an example of the free-rider problem. The argument that collectively optimal outcomes, when realized, are

34

realized because they are collectively optimal is one of the most frequent forms of functional explanation. It has two closely related defects, in that

it lacks both micro foundations and an appropriate feedback mechanism.

DIALECTICS

"Dialectics" is a term that has been used with a number of meanings. Common to almost all is the view that conflict, antagonism, or contradiction is a necessary

condition for achieving certain results. Contradiction between ideas may be a condition for reaching truth: conflict among individuals classes, or nations

may be a necessary condition for social change. This preliminary remark suggests a distinction between a dialectical method and a dialectical process,

between dialectics as a feature of our thinking about the world and dialectics as a feature of the world itself. On a certain conception of dialectics,

these are not alternative conceptions but complementary ones. The dialectical method reflects the dialectical character of the world. Hegel apparently

believed, at least some of the time, that our views about the world have to be contradictory because the world itself contains contradictions. This view

is hardly intelligible, and I shall not discuss it here.

Let us consider, therefore, the view that these are alternative conceptions. We may then define the dialectical method as the view that in order to arrive

at the truth of a matter one does not proceed by slowly and patiently refining earlier conceptions, deleting what is wrong, retaining what is correct,

and adding what is missing Rather, one goes from one extreme to another, discarding what is valuable in the old view along with what ought properly to

be discarded. In a third stage one may be able to achieve a more balanced view - but only because one has passed through the extremes.

William Blake, an older contemporary of Hegel, expressed this view in two succinct phrases: "Without Contraries is no Progression" and "You never know what

is enough unless you know what is more than enough." Hegel used different language. Sometimes he characterized the three stages as, respectively, thesis,

antithesis

35

and synthesis. In a different but essentially equivalent terminology he referred to them as position, negation of the position, and negation of the negation.

The Hegelian terms have a fine ring to them, but the phenomena they refer to are only the commonplace ones just described. Even if Hegel wanted to advocate

a special "dialectical logic," what remains valuable in his view can be expressed in everyday language and logic. To be valuable, however, it should not

be understood as asserting that this three-stage process characterizes all thinking, or that it is likely to yield better results than other procedures.

Intellectual development does not always proceed from one extreme to another, and when it does truth need not benefit

Dialectical processes in the world have similar stages. The most important example of a dialectical process in Hegel and Marx is probably the following

three-step sequence, briefly mentioned earlier. Society, they argued, begins as a primitive undifferentiated community. Persons are essentially similar

to one another with out distinctive character traits or different productive functions. The communitv dominates the individual, who is left with little

scope for free choice or individual self-realization. The next stage, the negation of the first one, occurs with the emergence of alienation (Hegel) or

of class societies (Marx). It is characterized by an extreme development of individuality and by an equally, extreme disintegration of community. The third

stage, the negation of the negation, restores commmunity without, however, destroying in- dividuality. It is in this respect the synthesis of the two previous

stages.

This vision had a very powerful grip on Marx's mind. It shaped his view of world history, his conception of political struggle and tactics, his image of

the communist society. It is related both to methodological holism and to functional explanation, sharing, therefore, the flaws of both. In other cases,

however, three-stage dialectical processes are less controversial. Consider, for instance, the development from the naive religious belief of the child,

through a stage of doubt and despair, to the reflective belief of the mature person. It is certainly arguable that the direct passage from the first to

the third stage is impossible. Even more plausibly, there

36

is no going back from the third stage to the first. If we regard these as the salient features of the negation of the negation, it appears as a common although

far from universal pattern. There is no "law" of the negation of the negation, but the concept has a certain value in directing our attention to problems

we might otherwise have overlooked.

There is no real connection between negation of the negation as a feature of thought processes and negation of the negation as a feature of historical processes.

There is no reason, that is, why the study of a three-stage dialectical process is more likely than other intellectual developments to proceed in three

dialectical stages. In any case I insist on the very limited interest of dialectics - in thought or in reality - as conceived in this manner. It does not

yield an operational method that can be applied with a promise of good results within well-defined boundaries, nor does it yield substantive laws of historical

development with definite predictions for concrete cases. A cluster of vague, suggestive ideas, it does not offer scientific tools with analytical cutting

edges.

There is, however, another acceptance of the term "dialectics" in which it does offer exactly that. It involves taking seriously, indeed literally, the

idea that the world contains contradictions. To see that this view does not also involve us in contradictions, consider the following propositions:

1. John Smith believes that it rains.

2. John Smith believes that it does not rain.

3. John Smith does not believe that it rains.

If we assert propositions 1 and 2 simultaneously, we make a statement to the effect that the world contains contradictions. If we assert 1 and 3 simultaneously,

we make a contradictory statement. This shows that a statement asserting